A review on efficient EEG pattern recognition using machine learning and deep learning methods and its application

摘要

With the good development of robotic technology, good communication of robots regarded as the most sought after achievement by researchers these days. If the robot can identify the feelings and intentions of the contact person, which can lead to more robots useful. Electroencephalography (EEG) is considered one of the most effective recording methods the emotions and motives of the brain user. The various types of machine learning are as follows used successfully to separate EEG data accurately. K-closest neighbor, Besesi network, Artificial Neural and Support Vector Machine networks are within the appropriate machine learning methods for classifying EEG data. The purpose of this concept is to explore different machine learning techniques to differentiate EEG data associated with specific emotional / emotional conditions. Different ways based on differences signal-processing techniques are studied Different numbers of EEG data elements are used to identify those that give the best results different classification techniques. Various methods are designed to format the database for EEG data. Formatted data sets were tested in various machine learning techniques in order to find out which process can place EEG data accurately according to it emotional / emotional conditions. On the other hand, this study is to study the electronic learning methods by various EEG tools. Continuous EEG is an exciting approach to cerebral function performance testing in intensive care unit and more. A systematic approach of this work derives various important outcome on EEG.